Explainable Machine Learning Model to Preoperatively Predict Postoperative Complications in Inpatients With Cancer Undergoing Major Operations.

PURPOSE: Preoperative prediction of postoperative complications (PCs) in inpatients with cancer is challenging. We developed an explainable machine learning (ML) model to predict PCs in a heterogenous population of inpatients with cancer undergoing same-hospitalization major operations. METHODS: Consecutive inpatients who underwent same-hospitalization operations from December 2017 to June 2021 at a single institution were retrospectively reviewed. The ML model was developed and tested using electronic health record (EHR) data to predict 30-day PCs for patients with Clavien-Dindo grade 3 or higher (CD 3+) per the CD classification system. Model performance was assessed using area under the receiver operating characteristic curve (AUROC), area under the precision recall curve (AUPRC), and calibration plots. Model explanation was performed using the Shapley additive explanations (SHAP) method at cohort and individual operation levels. RESULTS: A total of 988 operations in 827 inpatients were included. The ML model was trained using 788 operations and tested using a holdout set of 200 operations. The CD 3+ complication rates were 28.6% and 27.5% in the training and holdout test sets, respectively. Training and holdout test sets' model performance in predicting CD 3+ complications yielded an AUROC of 0.77 and 0.73 and an AUPRC of 0.56 and 0.52, respectively. Calibration plots demonstrated good reliability. The SHAP method identified features and the contributions of the features to the risk of PCs. CONCLUSION: We trained and tested an explainable ML model to predict the risk of developing PCs in patients with cancer. Using patient-specific EHR data, the ML model accurately discriminated the risk of developing CD 3+ complications and displayed top features at the individual operation and cohort level.

Pandemic Operating Room Supply Shortage and Surgical Site Infection: Considerations as We Emerge from the Coronavirus Disease 2019 Pandemic.

BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic created shortages of operating room (OR) supplies, forcing healthcare systems to make concessions regarding "standard" OR attire. At our institution, we were required to reduce shoe covers, reuse face masks, and allow washable head coverings. We determined if these changes affected surgical site infection (SSI) rates. STUDY DESIGN: A single institutional study was performed to compare the SSI rates reported to the National Healthcare Safety Network in the 2 years preceding COVID-19 (PRE, January 1, 2018, to December 31, 2020) with the first 12 months after the pandemic (POST, April 1, 2020, to March 31, 2021). We confirmed our findings using propensity score matching and multivariate analysis. RESULTS: Elimination of traditional shoe covers, disposable head covers, and single-use face masks was associated with a decreased SSI rate from 5.1% PRE to 2.6% POST (p < 0.001). Furthermore, this was despite a 14% increase in surgical volume and an increase in the number of contaminated/dirty cases (2.2% PRE vs 7.4% POST, p < 0.001). Use of disposable face masks decreased by 4.3-fold during this period from 3.5 million/y PRE to 0.8 million/y POST. Of note, inpatient hand hygiene throughout the hospital increased from 71% PRE to 85% POST (p < 0.001). CONCLUSIONS: This analysis has practical applications as we emerge from the pandemic and make decisions regarding OR attire. These data suggest that disposable head covers and shoe covers and frequent changes of face masks are unnecessary, and discontinuation of these practices will have significant cost and environmental implications. These data also reinforce the importance of good hand hygiene for infection prevention.